The role of donors, beyond the negative a priori

When looking at ownership issues in LICs, we must be aware of our own normative choices or biases. For political, moral, historical or empirical reasons, there is a temptation to formalise ownership by governments as the mirror reflection of the involvement of donors. The interaction between these two parties is then put as a ‘zero-sum game’ (what one would have, the other has not). Such a formalisation puts aid actors in a negative role. The view developed in this paper is that external actors can positively contribute to the success of health financing policies. As stated by Mayaka Ma-Nitu et al 9 in their contribution to the PBF debate, ‘exogeneity can raise problems in terms of sustainability, but this is far from axiomatic’. We do not discard the high sensitivity of the donor’s role or the ‘power imbalances’ between them and their country counterpart.10 11 We therefore believe that it is crucial to understand how they contribute, positively or not, to different stages of policy processes, for instance, through their control on financial resources, their (supposed) greater technical expertise and their privileged access to politicians.12

We acknowledge that aid actors have obligations for accountability (eg, towards tax payers from rich countries) and that these obligations will inevitably inform their involvement in recipient countries’ policies development. Our own perspective, (and the normative stance taken in this paper) is that a ‘balanced relationship’ between aid actors and their country counterparts is possible and desirable. It is important that the relationship ensures that aid actors, rather than obstructing (eg, by abusing their position), contribute to governmental and even better country ownership.

We argue that it is possible to progress towards this ideal and that there is probably more than one pathway to this goal. We will build on both our research on heath financing policies, including user fee removal initiatives,13 14 health equity funds15 16 and PBF,17–20 and our personal involvement in several countries depending on aid for their health sector (mainly Chad and Burkina Faso for JAK and Cambodia, Rwanda and Burundi for BM). We also tap on the vast literature on health financing schemes, with references to other strategies such as vouchers and health insurance.

Critical distance and knowledge as key resources for health financing policies

Central to health financing reforms is their technical nature. For any ambitious financing reform, one will have to answer some difficult questions on how to mobilise resources, how to organise their pooling and how to allocate them smartly, equitably and efficiently. The reform will often imply some substantial institutional reengineering and even some paradigm shift—these require both a readiness to take critical distance from the existing system and a strong knowledge base, especially with a view to ownership.

This is true even for policies which may look very straightforward like user fees removals. For the population, ‘free care’ is a clear concept and one that is easy to claim. However, in some countries, the health administration, some donors and even some public opinion, for both good and bad reasons, have sometimes been quite reluctant to remove user fees.21 22 Oftentimes, the necessary critical distance to deem user fees as a burden for the population came from incumbent presidents or candidates running for office—other lay knowledge holders. They have found out the possible political benefits of abolition of user fees and their own possible role in such a venture. This encounter between the citizens and the political leaders is not surprising and is to be welcomed, as it generates broad ownership in the society and is part of the social contract.

But critical distance is not enough: it needs to be backed with the required knowledge base. In the eyes of a president, user fee removal is relatively ‘easy’ to implement: you just need an abolition decree.23 But multiple country experiences have taught us that things are, of course, not so simple.24 There are several implementation challenges, including appropriate budgetary allocations, a clear definition of the benefit packages, a sufficient and timely ordering of drugs and the right provider payment system.25 26 Often, technical staff have been involved too little and/or too late; and limited attention and learning have been granted to the required financial reengineering. Today, some country health systems continue to suffer from decisions which have sometimes been hasty from a technical perspective. So, both knowledge and critical perspective are pivotal resources for health financing reforms.

The added value of involving external actors

In limited-resource settings, adopting, running and critically reflecting on health financing reforms are certainly possible without external actors (eg, Ghana's experience27 or the subsidy for obstetric care in Burkina Faso).28 Yet, health administrations in many countries do not always have the resources, expertise and maybe even the grit to reform the health system financing. This creates space for contribution by other actors, including external partners and experts who could be an asset regarding bringing critical perspective and expertise.

Indeed, once one adopts a knowledge-to-policy perspective, external actors (eg, non-governmental organisations (NGOs), bilateral aid agencies) have at least two advantages. On the one hand, they can serve as knowledge brokers: through their activities in other countries, they may have developed advanced technical expertise on the new funding mechanism to be tested or introduced. At least, they can help in finding and attracting relevant international experts who hold the required explicit and tacit knowledge (a group increasingly including experts from LICs). On the other hand, the aid instrument they typically use, the project, is particularly suited for experimentation, a key mechanism for both challenging possible dogmas and building a local knowledge base.29

The project format has its constraints (eg, sometimes too rigid), but it allows external actors to test new ideas, to introduce local changes without having to review the institutional system. It also allows governments to limit the adverse effects and political cost that could arise from failure to a small geographical area. Another advantage of the project format is that it allows external actors to address the very operational challenges (often related to knowledge building and diffusion) raised by the introduction of a health financing innovation: public budgets in LICs rarely have the flexibility required for funding of co-creation workshops, hiring of specialised expertise, development of guidelines, training on how-to-do things, development of software and so on.

Over the last two decades, in many LICs, a large array of health financing policies have been ‘incubated’ in projects supported by external players: health equity funds,15 30 vouchers,31 32 direct facility funding,33 contracting,17 34 35 PBF,19 36 community-based health insurance37 and even user fee removal.38 This approach has its drawbacks (eg, the never ending practice of experiments in some countries, insufficient early attention to scalability, the reproduction of schemes which have proven incompatible with universal health coverage (UHC)), but one must appreciate its main motive: all these schemes or policies are attempts to financially re-engineer health systems. They are often complex mechanisms and their design and implementation are not straightforward.18 27 39–41 Given the heavy stake of acquisition of new knowledge, a pilot phase has often been adopted. It has allowed country actors to learn and answer questions such as: is this mechanism suitable for our country? what are the necessary adaptations for a good fit with our local context?

It is important to keep in mind that often, the entire health system has to ‘discover’ the mechanism and get more acquainted with it.42 Indeed, most of these financing mechanisms are not only new to the country, they are also often exogenous to the 'health pyramid' itself: they challenge the existing organisational system, they adopt a purchaser-provider split logic and require a new actor ‘external' to the health pyramid: the purchaser.43 Thus, ownership is a matter of attention.

Ownership: which one and when does it really matter?

Our own practice taught us that ownership of a health financing innovation should be shared across stakeholders, whatever the pathway taken: there are so many actors who hold some piece of the knowledge that forgetting some could be detrimental. So, beyond governmental ownership, we believe that it is national (or country) ownership (including possibly actors as diverse as Presidency, Prime Ministry, Ministry of Finance, Ministry of Health, other relevant Ministries, local governments, parliaments, municipalities, front line workers, private sector, civil society organisations, academia, etc) which matters the most.

But at what stage in the policy process should ownership be judged? Considering governmental ownership, Gautier and Ridde7 proposed different stages of a heuristic framework. Bertone et al 44 adopted a similar view in their study on PBF in fragile and conflicted affected countries. While adopting a sequential approach is probably the right way to look at the emergence of the national ownership,45 we believe that what really matters in terms of national ownership is whether the policy reaches the status of being abided by all relevant parties as a new institutional arrangement in the intended jurisdictional area (country or state in federal countries).46

In fact, whatever the degree of involvement and the role played by external actors, it is impossible for a health financing policy to be established nationwide and certainly to continue for several years without broad national ownership. Actually, the policy reaches an ‘equilibrium’, a term used in microeconomics to describe a state from where no one has an interest to move away. This observation applies to, for instance, the health insurance policy in Ghana,27 the subsidy for obstetric care28 or the user fees removal policy in Burkina Faso,38 the Régime d’Assistance Médicale (Medical Assistance Plan) in Morocco,47 the mutuelles in Rwanda48 or the PBF-Free Healthcare in Burundi.49 50 There is ownership when key stakeholders see clear value in the policy and will oppose its dismantlement. This predictable opposition implies a political cost for national authorities, including a possible setback at the next election and sometimes, even a reputation risk for the country. This can of course also create a challenge for reforming the strategy itself—but this is what ownership is about: strong and conscious support.

More than one pathway to achieve country ownership

When we combine the recognition that (1) policy is a process whose most important moment is the stable state when the policy is valued and therefore abided by all relevant stakeholders and (2) the fact that ownership can be shared across them, we are creating space for a much richer discussion of how broad societal support can be developed for a policy. Ourselves, we have witnessed a variety of pathways.

A first pathway could be categorised as ‘technicians first, politicians second’. Under this scenario, a small group of actors (eg, senior officials from ministries, academics, national/international technical assistants), after having prepared the ground by informing the proper authorities and obtaining the necessary green lights, use their decision space to test a new mechanism at small scale, off the national budget.19 51 Once they have a proof of concept, they will act as a coalition of policy entrepreneurs52: they will share their results and seek buy-in from ministers and the government to induce a scale-up, access the public budget and obtain national policy status. Quite often this dialogue is strengthened with donors’ support. Obviously, if some of the policy entrepreneurs have personal connections with key politicians or have the ‘right’ political affiliation, it helps the policy dialogue. This is the path Rwanda followed, both for its mutuelles and PBF policies. We have extensively described this trajectory in several countries in a recent study on Results-Based Financing policies.45 53 Among other things, we have shown that the composition of the supportive coalition is evolving across policy stages. We also know that this trajectory can fail, for instance because of the lack of strong national policy entrepreneurs or the high turnover of key players such as Ministers or senior officials, in which case the embryonic policy dies.54 In general, the implementation of these policies is gradual, especially when the political payoff is not obvious.

A second pathway could be categorised as ‘politicians first, technicians second’. This is a scenario which has been followed for most user fee removal experiences. The removal is owned by the national authorities, but a lack of early involvement of some key national technicians and health staff (so-called ‘street-level bureaucrats’55) in the policy sometimes creates ownership problems at the implementation stage.7 As for the other pathway, the final outcome may be positive (a long term institutionalisation of the policy) or more negative (policy goes on for a while still, but its defects affect its effectiveness and its collapse may turn out to be one long agony). Actually, in this scenario, the political will is strong from the outset—policies often start with a bang and required accompanying reforms get the necessary political support.

A third pathway could be a close and inclusive interaction, from a very early stage, between the political and technical levels—something certainly highly recommendable as such co-production will not only allow taking into account both the technical and political perspectives, but also ensure that individuals who contributed to the process develop a strong bond with the reform (this will be an asset when the policy needs to be explained or defended). This is probably the ideal way to progress on health financing policy design and implementation. Obviously, the development and implementation of the National Health Insurance Scheme in Ghana,56 the obstetric subsidy and the user fee removal policy in Burkina Faso,28 and the free caesarian section initiative in Benin57 were made possible this way. The story of Burundi which managed to integrate, in 2010, the ‘politicians first, technicians second’ user fee removal for children under-five and deliveries into the ‘technicians first, politicians second’ PBF is also very interesting, as this strategy allowed the initiative to secure, eventually, support from both politicians and technicians.49 50

Are there other pathways? Definitely. If we look for instance at the role that civil society organisations (CSOs) could play, especially those ensuring a watchdog function of the health system. Indeed, thanks to their critical perspective, they note dysfunctions or ‘undesirable situations’; they can then alert public authorities, sometimes with solutions in hand to address them. In practice, CSOs may need support or collaboration from technicians or peers to refine and/or push their solution. This was the case in Burkina Faso with the local NGO Réseau d’Accès aux Médicaments Essentiels (Access Network to Essential Medicines) whose watch activities have enabled the nation-wide user fee removal for HIV care, including antiretroviral drugs and some diagnostic tests.

The way forward

We hope that this broad reflection on how ownership can be facilitated during policy processes will help policy makers, aid agencies, researchers and other stakeholders in their future actions.

The fact that a growing number of international actors promote approaches securing a more proactive involvement of country expertise, multistakeholder coproduction and regional collaboration is welcome.58–60 These could be achieved through mechanisms as diverse as peer-to-peer exchanges, communities of practices, public expertise sharing, mentoring and coaching. This should contribute to a better handling of the technical issues which matter for the policy ‘equilibrium’, prevent major flaws in policy design and implementation, and thus ensure that the policy has key desirable traits (efficiency, equity, transparency and accountability).

But external and national actors have to recognise that health financing is not just a technical agenda: it is a major political issues for entire countries.61 So, enough attention should also go to issues which can threaten the ‘equilibrium’ from a more political perspective: is the policy generating enough political pay-off for the government; does it meet the reasonable expectations of health staff; is there a consolidation of the evidence base in terms of outcomes; is there the required fiscal space? Several of the experiences mentioned above suggest the existence of links between these issues. For instance, the visibility of the entitlement granted to the citizens will partly determine the sustained fiscal commitment from the highest national authorities.

But for many of these questions, external actors’ contribution will be secondary: the primary role is for country actors themselves. Indeed, what matters most is the capacity to build and sustain democratic accountability that could be defined as ‘the many ways in which citizens, political parties, parliaments and other democratic actors can provide feedback to, reward or sanction officials in charge of setting and enacting public policy’.62 It is plausible that health financing policies are not equivalent in this respect. Some pathways could be more conductive than others to such democratic accountability, and thus to the policy ‘equilibrium’. One can assume that UHC benefits from democracy, as national authorities have an electoral mandate to care for the health demands and needs of their population.63 While it is true that in many LICs, a real democratic process is struggling to settle, things are improving in others, leading to optimism. In the face of such changes, new skills, attitudes and practices should be developed by external partners.

Understanding the mechanisms triggering buy-in from different stakeholders (especially national governments) and democratic accountability requires more attention. For understanding country realities, scientists can help: we need more theoretical and empirical research on the political economy of health financing reforms in LICs.42 In parallel, we should strengthen our support to governments to fulfil their obligations towards their own policies.